Function Theory on Planar Domains: A Second Course in Complex Analysis
()
About this ebook
The first chapter and parts of Chapters 2 and 3 offer background material, all of it classical and important in its own right. The remainder of the text presents results in complex analysis from the far, middle, and recent past, all selected for their interest and merit as substantive mathematics. Suitable for upper-level undergraduates and graduate students, this text is accessible to anyone with a background in complex and functional analysis. Author Stephen D. Fisher, a professor of mathematics at Northwestern University, elaborates upon and extends results with a set of exercises at the end of each chapter.
Related to Function Theory on Planar Domains
Related ebooks
Axiomatic Set Theory Rating: 4 out of 5 stars4/5Complex Analysis Rating: 3 out of 5 stars3/5Functional Analysis: Introduction to Further Topics in Analysis Rating: 5 out of 5 stars5/5Special Functions & Their Applications Rating: 5 out of 5 stars5/5Extremal Graph Theory Rating: 3 out of 5 stars3/5Singular Integrals and Differentiability Properties of Functions Rating: 0 out of 5 stars0 ratingsIntroduction to Combinatorial Analysis Rating: 5 out of 5 stars5/5An Introduction to Algebraic Structures Rating: 2 out of 5 stars2/5Topological Vector Spaces and Distributions Rating: 3 out of 5 stars3/5Introduction to Topology: Second Edition Rating: 5 out of 5 stars5/5Uniform Distribution of Sequences Rating: 4 out of 5 stars4/5Analysis in Euclidean Space Rating: 0 out of 5 stars0 ratingsApplied Functional Analysis Rating: 0 out of 5 stars0 ratingsAnalytic Functions Rating: 0 out of 5 stars0 ratingsAlgebraic Number Theory Rating: 0 out of 5 stars0 ratingsProbability: A Survey of the Mathematical Theory Rating: 0 out of 5 stars0 ratingsGeometry and Symmetry Rating: 0 out of 5 stars0 ratingsSet Theory: The Structure of Arithmetic Rating: 5 out of 5 stars5/5Number Theory Rating: 4 out of 5 stars4/5Counterexamples in Topology Rating: 4 out of 5 stars4/5Numerical Analysis Rating: 0 out of 5 stars0 ratingsOn Riemann's Theory of Algebraic Functions and Their Integrals: A Supplement to the Usual Treatises Rating: 0 out of 5 stars0 ratingsPartial Differential Equations: An Introduction Rating: 2 out of 5 stars2/5General Topology Rating: 4 out of 5 stars4/5Relativity for Scientists and Engineers Rating: 4 out of 5 stars4/5A Second Course in Complex Analysis Rating: 0 out of 5 stars0 ratingsA Course in Advanced Calculus Rating: 3 out of 5 stars3/5Stability & Periodic Solutions of Ordinary & Functional Differential Equations Rating: 0 out of 5 stars0 ratingsMathematics of Classical and Quantum Physics Rating: 3 out of 5 stars3/5Real Analysis: Measure Theory, Integration, and Hilbert Spaces Rating: 4 out of 5 stars4/5
Mathematics For You
What If?: Serious Scientific Answers to Absurd Hypothetical Questions Rating: 5 out of 5 stars5/5My Best Mathematical and Logic Puzzles Rating: 4 out of 5 stars4/5Algebra - The Very Basics Rating: 5 out of 5 stars5/5The Little Book of Mathematical Principles, Theories & Things Rating: 3 out of 5 stars3/5Alan Turing: The Enigma: The Book That Inspired the Film The Imitation Game - Updated Edition Rating: 4 out of 5 stars4/5Pre-Calculus For Dummies Rating: 5 out of 5 stars5/5Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways to Lie with Statistics Rating: 4 out of 5 stars4/5Basic Math & Pre-Algebra For Dummies Rating: 4 out of 5 stars4/5Calculus For Dummies Rating: 4 out of 5 stars4/5Quantum Physics for Beginners Rating: 4 out of 5 stars4/5Math Magic: How To Master Everyday Math Problems Rating: 3 out of 5 stars3/5Real Estate by the Numbers: A Complete Reference Guide to Deal Analysis Rating: 0 out of 5 stars0 ratingsRelativity: The special and the general theory Rating: 5 out of 5 stars5/5Basic Math & Pre-Algebra Workbook For Dummies with Online Practice Rating: 4 out of 5 stars4/5Sneaky Math: A Graphic Primer with Projects Rating: 0 out of 5 stars0 ratingsBasic Math and Pre-Algebra: 1,001 Practice Problems For Dummies (+ Free Online Practice) Rating: 3 out of 5 stars3/5Mental Math Secrets - How To Be a Human Calculator Rating: 5 out of 5 stars5/5Algebra I Workbook For Dummies Rating: 3 out of 5 stars3/5Algebra II For Dummies Rating: 3 out of 5 stars3/5The Golden Ratio: The Divine Beauty of Mathematics Rating: 5 out of 5 stars5/5Flatland Rating: 4 out of 5 stars4/5How to Solve It: A New Aspect of Mathematical Method Rating: 4 out of 5 stars4/5Logicomix: An epic search for truth Rating: 4 out of 5 stars4/5Calculus Made Easy Rating: 4 out of 5 stars4/5Intermediate Algebra Rating: 0 out of 5 stars0 ratingsGeometry For Dummies Rating: 4 out of 5 stars4/5Must Know Math Grade 7 Rating: 5 out of 5 stars5/5Trigonometry For Dummies Rating: 0 out of 5 stars0 ratings
Reviews for Function Theory on Planar Domains
0 ratings0 reviews
Book preview
Function Theory on Planar Domains - Stephen D. Fisher
Domains
Function Theory on
Planar Domains
A Second Course in Complex Analysis
Stephen D. Fisher
Department of Mathematics
Northwestern University
Dover Publications, Inc.
Mineola, New York
Copyright
Copyright © 1983 by Stephen D. Fisher
All rights reserved.
Bibliographical Note
This Dover edition, first published in 2007, is an unabridged republication of the work published by John Wiley & Sons, Inc., New York, in 1983.
Library of Congress Cataloging-in-Publication Data
Fisher, Stephen D., 1941–
Function theory on planar domains: a second course in complex analysis / Stephen D. Fisher.
p. cm.
Originally published: New York: Wiley, c1983, in series: Pure and applied mathematics.
Includes bibliographical references and index.
eISBN-13: 978-0-486-15110-6
1. Functions of complex variables. 2. Hardy spaces. I. Title.
QA331.F59 2007
515'.9—dc22
2006052123
Manufactured in the United States of America
Dover Publications, Inc., 31 East 2nd Street, Mineola, N.Y. 11501
To my family
NAOMI, KEFIRA, HANAH, and EFREM
PREFACE
Complex analysis is a member of the first generation of modern mathematics, and substantial portions of today’s mathematics can trace their lineage directly back to it. Despite its advanced age, however, the subject remains active and fruitful, death sentences from other quarters notwithstanding. It is my hope that this book will bear witness to this assertion and that each reader will find stimulation and excitement in seeing some of the significant results that complex analysis continues to produce.
The plan of this book is simple. The first chapter and some sections of Chapters 2 and 3 are background material, all of it classical and important in its own right. The remainder of the book presents results in complex analysis from the far, middle, and recent past and even the near present, results that I have selected because of their interest and their merit as substantive mathematics. The selection was necessarily made with my own prejudices and I have been forced to omit many other topics because of the natural limitations of size forced on any such effort. Anyone with a basic graduate course in complex analysis and in real variables-functional analysis has an adequate background to read, understand, and enjoy the topics in this book.
I have frequently used the exercises at the end of each chapter to elaborate on and extend results given in the text; the serious reader would be well advised to work the exercises since mathematics is best understood by doing, rather than only reading. Occasionally in the text there is an opportunity to prove something in two ways; in such cases I have usually opted for the less sophisticated proof, feeling that this was more in keeping with the spirit of the book.
The influence of my teachers at the University of Wisconsin—Frank Forelli, Walter Rudin, Simon Hellerstein, Michael Voichick, and Anatole Beck—is present throughout this book for they not only taught me complex analysis, they also proved many of the theorems to be found in this book. My colleagues at M.I.T. during the fruitful year 1967–1968—Ted Gamelin, John Garnett, Ken Hoffman, Don Wilken, and Larry Zalcman—also influenced me; our analysis seminar was always productive and informative, even if unappreciatedly dry. Many results of theirs are also to be found herein.
I have made some effort at the end of each chapter to give references to related work and to recount the history of the main theorems of that chapter; any omissions of references are inadvertent and regretted by me. In no case should a theorem be attributed to me due solely to an absence of other credits.
In a book of this kind there is inevitably some overlap with other books, especially on the standard
material, and I have made note of this in the comments at the end of the chapters. A major theme of this work is to elaborate on function theory on a finitely connected planar domain, to show in what ways it resembles and in what ways it differs from function theory on the disc, and to do this without the artifice of saying one can easily see that. . . .
Of course, there is a great deal more, as well, including a number of theorems on quite arbitrary domains.
This book evolved from courses I taught from time to time. Thanks are especially due to The Technion, Haifa, Israel, for providing me with the opportunity to teach a seminar in the spring of 1980 on much of the material in Chapters 1, 3, and 5. I also wish to thank T. W. Gamelin who read a first draft of the book and provided me with many helpful comments.
Finally, I write usually with the editorial we
not only because it seems awkward to always write in the first person singular but also because reading a mathematics book is a joint effort between author and reader and the latter is being escorted, as it were, by me through the thickets and meadows of an interesting landscape and we are seeing
things together.
STEPHEN D. FISHER
Evanston, Illinois
January 1982
CONTENTS
Notation and Numbering
1.The Dirichlet Problem and Harmonic Measure
1.1.Introduction
1.2.The Poisson Formula and Some Preliminaries
1.3.Subharmonic Functions
1.4.Solution of the Dirichlet Problem
1.5.The Green’s Function of a Domain
1.6.Harmonic Measure
1.7.Logarithmic Capacity
Additional Readings and Notes
Exercises
2.Uniformization and Conditional Expectation
2.1.Introduction
2.2.The Uniformization Theorem
2.3.Conditional Expectation and the Space N
Additional Readings and Notes
Exercises
3.The Hardy Spaces Hp(Ω)
3.1.Introduction
3.2.Basic Properties of Hp(Ω)
3.3.Hp on the Unit Disc
and Hp(Ω)
3.5.Null Sets and Essential Boundary Points for H∞(Ω)
3.6.H∞(Ω) Determines Ω*
3.7.Weak Peak Points for H∞(Ω)
Additional Readings and Notes
Exercises
4.Domains of Finite Connectivity
4.1.Introduction
4.2.The Defect of ReR(Ω) in Cr(Γ)
4.3.Measures Orthogonal to R(Ω)
4.4.Hp(Ω)
4.5.N Again
4.6.Functions with Periods
4.7.The Factorization of Hp(Ω) Functions
Additional Readings and Notes
Exercises
5.Blaschke Products, Inner Functions, and Extremal Problems
5.1.The Ahlfors Function
5.2.Blaschke Products
5.3.Approximation by Inner Functions
5.4.Pick-Nevanlinna Interpolation
5.5.Interpolation Sequences
5.6.The Maximum Principle for Multiple-Valued Bounded Analytic Functions
Additional Readings and Notes
Exercises
6.The Maximal Ideal Space of H∞(Ω)
6.1.Introduction
6.2.Peak Points and Parts
6.3.The Fibers of M(Ω)
6.4.Distinguished Homomorphisms
6.5.The Shilov Boundary of H∞(Ω)
6.6.The Corona Theorem
Additional Readings and Notes
Exercises
7.Linear Operators on Hp Spaces
7.1.The Isometries of HP(Ω)
7.2.Self-Mappings of a Domain
7.3.General Properties of Composition Operators
7.4.Compact Composition Operators on H∞(Ω)
7.5.Optimal Estimation and Widths of Spaces of Holomorphic Functions: Part 1. The H∞ Case
7.6.Optimal Estimation and Widths of Spaces of Holomorphic Functions: Part 2. The H² Case
Additional Readings and Notes
Exercises
Bibliography
Index
NOTATION AND NUMBERING
In this book, we make use of several notations consistently and thus it is worthwhile to list them here.
Δ,the open unit disc = {z: |z| < 1}
T,the unit circle = {z: |z| = 1}
σ,normalized Lebesgue measure on T: dσ = (1/2π)dθ
S², the Riemann sphere
; Swith the point
∞ attached.
or in S²; that is, an open connected set.
For a set E, CL(E) is the closure of E and Ec is the complement of E, relative to S². Further, ∂E is the boundary of E and INT E is the interior of E. Finally, C(E) is the space of continuous complex-valued functions on E, Cr(E) is the subspace of real-valued continuous functions, and ||u||E is the supremum of ||u(x)| as x varies over E.
A measure is always a finite regular Borel measure.
A word on the numbering system in this book is also in order. Within each section, say Section 2 of Chapter 3, the propositions and theorems are numbered consecutively using a system of double arabic numbers: 2.1, 2.2, and so on, and reference to them from within that chapter is made accordingly. However, if in another chapter reference is made to, let’s say, Theorem 2.2 of Chapter 3, then we write see Theorem 3.2.2.
In this system of triple arabic numbers, the first number refers to the chapter in which the proposition or theorem is found. This should cause no difficulties. A similar convention applies to numbered formulas.
Function Theory on
Planar Domains
1
THE DIRICHLET PROBLEM AND HARMONIC MEASURE
1.1.INTRODUCTION
Let Ω be a domain on the Riemann sphere and let u be a continuous real-valued function on Γ = ∂Ω. The Dirichlet problem is to find, if possible, a function f which is continuous on Ω ∪ Γ = CL(Ω) and which satisfies the following conditions:
1.f is harmonic on Ω; that is, Δf = 0 on Ω.
2.f = u on Γ.
There certainly are cases in which this problem is not solvable; for example, if Ω = {z: 0 < |z| < 1} and u. However, there are a wide variety of domains for which it is solvable since there are quite reasonable conditions that are sufficient for solvability. The standard approach is by the method of Otto Perron and makes use of subharmonic functions. We begin in Section 2 with some material on the Poisson integral and proceed to the definition of subharmonic functions and some of their basic properties in Section 3; in Section 4, we use subharmonic functions to solve
the Dirichlet problem. Related matters, including harmonic measure, the Green’s function, and logarithmic capacity are covered in Sections 5, 6, and 7.
1.2.THE POISSON FORMULA AND SOME PRELIMINARIES
It is worthwhile to begin by recalling the Poisson integral formula and some of its basic properties. Set
P(r, θ) is the Poisson kernel and it is a simple matter to verify that
Now let u be a continuous function on the unit circle T, and set
The function Pu is a harmonic function of z = reit as (2.2a) shows. The significant thing about Pu is what Pu(z) does as z tends to a point of T.
Theorem 2.1. Pu(z) → u(λ) as z → λ, λ ∈ T; that is, Pu is continuous on Δ ∪ T and coincides on T with u.
Proof. The proof is the standard application of an approximate identity argument and makes use of (2.2b), (2.2c), and (2.2d). Let λ = eiψ. Then
Given ε > 0, choose δ > 0 so that whenever |θ – ψ| < δ it follows that |u(eiθ) – u(eiψ)| < ε/2; from now on we view t as being restricted by |t – ψ| < δ/2. Next, for this δ, let rit follows that
. In the preceding identity for Pu(reit) – u(eiψ) we estimate the first integral by ε/2 since (2.2b) and (2.2c) hold. We estimate the second integral by
because of (2.2b), (2.2d) and (2.3). Thus, the distance from Pu(reit) to u(eiψ) is at most ε when |ψ – t| < δ, which is precisely what was to be shown.
Definition. Let μ be a measure on T and set
The function Pμ, which is harmonic in Δ, behaves relatively well at the unit circle T.
Theorem 2.2. Let dμ = υ dθ + dα be the Lebesgue decomposition of μ where υ ∈ L¹(T, dθ) and dα is singular with respect to dθ. Then
Proof. This proof is like that of Theorem 2.1 but, since we are only looking at radial limits, it is somewhat simpler. The measure dμ is given by the function μ of bounded variation on [–π, π] and (2.5) will follow if we show that
at each point θ0 at which μ is differentiable. Note that if dμ = 1/(2π) dθ, then P ≡ 1 and so (2.6) holds. Hence, we may subtract a constant multiple of 1/(2π)dθ from dμ and so assume that ∫dμ = 0; that is, μ(–π) = μ(π).
One integration by parts leads to
Now P′(r, ψ) = [2r(1 – r²)sin ψ]/[(1 – 2r cos ψ + r²)²] and is an odd function of ψ. Thus, we obtain
The functions
form an approximate identity; that is, (2.2b)—(2.2d) hold with K(r, t) in place of P(r, t). The function
is continuous at t = 0 with value F(0) = μ′(θ0). Hence
which is the desired conclusion.
Proposition 2.3. A harmonic function u in Δ has the representation
for some measure μ on T if and only if
If (2.7) holds, then μ is uniquely determined.
Proof. Suppose (2.7) holds. Then
Here we have made use of (2.2b) and (2.2c). Conversely, suppose (2.8) holds. Let μρ be the measure on T given by
so that (2.8) says exactly that the total variation of μρ is uniformly bounded for 0 < ρ < 1. There thus is a measure μ on T which is a weak-* cluster point of {μρ}. Hence,
As for uniqueness, suppose μ1 also satisfies (2.7). The difference μ – μ1 then annihilates all P(r, θ – t) and we must show that this implies μ – μ1 is zero. By taking real and imaginary parts we need only show that if ν is a real measure with
then ν = 0. From (2.9) and (2.2a) we have
so that the analytic function
is identically constant and hence 0 since h(0) = 0. However,
Hence,
Since ν is real, this implies ν is the zero measure.
Corollary 2.4. If u is a positive harmonic function on Δ then there is a unique non-negative measure μ on T with
Proof. Since u is positive it satisfies (2.8):
Thus, (2.10) holds for some measure μ; note that Proposition 2.3 actually yielded the information that
In this case, this supremum is u(0); hence
so that μ is non-negative.
1.3.SUBHARMONIC FUNCTIONS
Definition. A function u(z) defined for z in a domain Ω on the sphere is subharmonic on Ω if it satisfies the following conditions:
u is upper semicontinuous on Ω; that is,
lies in Ω, then
Evidently every real-valued harmonic function on Ω is subharmonic and if both u and –u are subharmonic then u is harmonic, since in this case u will be continuous and equality will hold in (3.1c). It is also evident that the sum and the maximum of two subharmonic functions are also subharmonic, as is a positive multiple of a subharmonic function. We gather some simple facts about subharmonic functions in the next several propositions.
Proposition 3.1. Let u be subharmonic on Ω and let ϕ . Then ϕ(u(z)) is subharmonic on Ω.
Proof. If we set υ(z) = ϕ(u(z)) then it is apparent that υ satisfies (3.1a) and (3.1b). Further, we have
since ϕ is both increasing and convex.
EXAMPLE.
Let f be holomorphic on Ω. Then both log| f | and | f |q, 0 < q < ∞, are subharmonic on Ω.
This is reasonably straightforward. Clearly (3.1a) and (3.1b) are satisfied. Further, (3.1c) is direct for | f |q since
by Hölder’s inequality. However, this argument won’t work for 0 < q < 1, so we first show log| f | is subharmonic and then apply Proposition 3.1 with ϕ(t) = eqt.
is in Ω; we assume first that f ≠ 0 on |z – p| = r. Let z1, . . ., zN be the zeros of f in |z – p| < r, and put
Then g , | g(z)| = | f(z)| if |z – p| = r, and | g(z)| > | f(z)| if |z – p| < r. Further, g so that log |g| is harmonic there. Thus,
Finally, if f does vanish on |z – p| = r choose rn ↑ r so that f ≠ 0 on |z – p| = rn.
Thus,
which establishes the desired inequality.
Proposition 3.2. Let {un, z ∈ Ω and limn→∞ un(z0) = L > – ∞ for some z0 ∈ Ω. Then u(z) = limn→∞ un(z) is subharmonic on Ω.
Proof. Suppose that p ∈ Ω is a point at which – ∞ < u(plies in Ω, then
But the latter integrals converge to the integral of u by the monotone convergence theorem. Hence, u satisfies (3.1c) at p. Moreover, u(p + reit) > – ∞ for almost all points (dt) on the circle |z – p| = r.
Next let a be any point of Ω at which u(a) > – ∞. Then, given ε > 0, there is a large n for which u(a) > un(a) – ε. Thus,
and so u is upper semicontinuous at a. If u(a) = – ∞, then given a large positive number M, we know that un(a) < –M for all large n. Hence,
so that lim supz→a u(z) = – ∞, in the case u(a) = – ∞.
Proposition 3.3. Suppose u ∈ Cin Ω, then u is subharmonic on Ω.
Proof. lies in Ω. Green’s theorem gives
However, the first integral is non-negative so that
is an increasing function of t. As t → 0, this integral converges to u(p), by continuity. Hence, u satisfies (3.1c).
We need a simple fact about upper semicontinuous functions which we isolate here; its proof is left as an exercise.
Lemma 3.4. Let K be a compact set and let u be a function on K with values in [– ∞, ∞). Then u is upper semicontinuous if and only if there is a sequence {fn} of continuous functions on K and lim fn(z) = u(z), z ∈ K.
The most important fact about subharmonic functions and the origin of their name is contained in the next result.
Theorem 3.5. An upper semicontinuous function u on Ω with values in [– ∞, ∞) is subharmonic if and only if, whenever K is a compact subset of Ω and h is a function continuous on K and harmonic on INT K with on ∂K, then on INT K as well.
Proof. One direction of the theorem is straightforward. Suppose u lies in Ω, then let {fn} be a sequence of continuous functions on the compact set γ = {z: |z – p| = r) which decrease to u. Again denote by fn , harmonic on {z: |z – p| < r) which equals fn on γ inside γ as well, by the hypothesis. Thus,
by the monotone convergence theorem. Thus, u satisfies (3.1c).
Conversely, let u be subharmonic in Ω, let K be a compact set in Ω and let h be continuous on K, harmonic on INT Kon ∂K. Set υ = u – h on INT K. If not, let
and
Since υ is upper semicontinuous, E on ∂K, we also know that E is a subset of INT K. Let p be a point in ∂E and choose r lies in INT K. Then on some arc of the circle γ = {z: |z – p| = r, δ . But then we obtain the contradiction
in INT K, as desired.
Proposition 3.6. If u ∈ C²(Ω), then u on Ω.
Proof. Suppose u , let υ be the harmonic function on |z – p| < r2 which agrees with u on |z – p| = rin |z – p| < r2 so that
is an increasing function of t and so
in Ω.
There is a maximum modulus
theorem for subharmonic functions.
Proposition 3.7. Suppose there is a number M < ∞ such that
for all z ∈ Ω. If u(z0) = M for some z0 ∈ Ω, then u ≡ M in Ω.
The proof is virtually the same as that of Theorem 3.5 and we leave it as an exercise.
As a finale to this section we note that subharmonicity is preserved by conformal maps, a result we will have need of later. Specifically, let ϕ be a one-to-one holomorphic mapping of a domain Ω onto a domain ′Ω1 and suppose u1 is subharmonic in Ω1. Then u(z) = u1(ϕ(z)) is subharmonic in Ω, as is easily verified by use of Theorem 3.5.
1.4.SOLUTION OF THE DIRICHLET PROBLEM
The fundamental result needed to attack the Dirichlet problem is this.
Proposition 4.1be a family